In a move aimed at curbing the spread of misinformation, TikTok announced plans to label images and videos uploaded to its platform that have been generated using artificial intelligence (AI) technology.

The video-sharing service will employ a digital watermarking system known as Content Credentials to identify AI-generated content.

The decision comes amid growing concerns from researchers that AI-generated content could be weaponised to interfere with the upcoming US elections. Earlier this year, TikTok was among 20 tech companies that signed an accord pledging to combat AI-powered misinformation.

While TikTok already labels AI-generated content created within its app, the new measure extends to videos and images produced outside the service using AI tools. "We also have policies that prohibit realistic AI that is not labelled," said Adam Presser, TikTok's head of operations and trust and safety. "If realistic AI-generated content appears on the platform, we will remove it as a violation of our community guidelines."

The Content Credentials technology is an industry-wide initiative spearheaded by the Coalition for Content Provenance and Authenticity, co-founded by Adobe, Microsoft, and others. Major players like ChatGPT maker OpenAI, YouTube, and Meta have also adopted the technology.

The system works by having AI tool developers, such as OpenAI, embed a digital watermark and tamper-proof data into the generated content. When such marked content is uploaded to platforms like TikTok, it is automatically labelled as AI-generated, provided the platform supports the Content Credentials standard.

TikTok's move comes as the company faces heightened scrutiny from US authorities. With 170 million users in the country, TikTok is owned by China's ByteDance, which was recently ordered to divest the app or face a ban under a new US law. ByteDance has sued to block the law, arguing it violates the First Amendment.

---